At Databricks, I build Python-based autonomous enterprise AI agents and RAG systems serving more than 165,000 monthly users. My work improved response accuracy by 34% and reduced response latency by 41% across knowledge-intensive workflows.
I design agent orchestration for tool calling, multi-step planning, vector retrieval, LLM reasoning, and structured execution. I also optimize model routing, prompts, caching, and vLLM serving to reduce GPU inference costs by 29%.
I build cloud-native AI services with FastAPI, Docker, Kubernetes, AWS, MLflow, and distributed data pipelines using Spark, Delta Lake, and vector indexing. I integrate enterprise platforms including ServiceNow, Jira, Slack, Workday, and Salesforce through secure APIs and event-driven microservices.
Previously at Accenture, I developed machine learning pipelines and real-time inference services for enterprise applications processing millions of daily events. I improved incident-prediction accuracy by 31%, reduced prediction latency by 42%, and supported scalable model deployment on AWS.
